A new review and perspective paper titled “Artificial Intelligence-Powered Raman Spectroscopy through Open Science and FAIR Principles” has been published in ACS Nano (DOI: 10.1021/acsnano.5c09165). Co-authored by researchers from CNR-IFN (Renzo Vanna and Victor Alcolea Rodriguez), the article outlines how Open Science and digital workflows can drive the integration of artificial intelligence into Raman spectroscopy.
Key Insights & The FAIR Paradigm
While Raman spectroscopy and chemometrics are advancing rapidly, data fragmentation and proprietary formats still restrict the full potential of AI. The authors emphasize that unlocking AI-driven workflows requires adopting the FAIR principles (Findable, Accessible, Interoperable, Reusable):
- Findable & Accessible: Using persistent identifiers (PIDs) and storing spectral datasets with rich metadata in open-access repositories.
- Interoperable & Reusable: Standardizing data structures (e.g., NeXus/HDF5), implementing ontologies like VIBSO, and sharing open calibration protocols and analysis pipelines.
Adopting these practices ensures high-quality, harmonized training datasets for machine learning models, mitigating issues like dataset shift, overfitting, and publication bias.
Synergy with the TROPHY Project
This publication aligns closely with the goals of the EU-funded TROPHY project (ulTRafast hOlograPHic FTIR microscopY, EIC Pathfinder, GA 101047137, www.trophy-project.eu). By developing novel label-free vibrational imaging techniques, TROPHY strongly relies on standardized data collection, advanced chemometrics, and open practices to ensure seamless data integration and reproducibility across modalities.
N. Coca-Lopez, V. Alcolea-Rodriguez, M. A. Bañares, S. Brockhauser, J. Gorenflot, A. Henderson, R. Hildebrandt, N. Jeliazkova, N. Kochev, E. Lozano Diz, Z. Pilat, D. Polli, P. Strömert, C. Sturm, R. Vanna, R. Portela. Artificial Intelligence-Powered Raman Spectroscopy through Open Science and FAIR Principles. ACS Nano 2025, 19, 38189–38218.
Read the article: https://doi.org/10.1021/acsnano.5c09165
Citation:
N. Coca-Lopez, V. Alcolea-Rodriguez, M. A. Bañares, S. Brockhauser, J. Gorenflot, A. Henderson, R. Hildebrandt, N. Jeliazkova, N. Kochev, E. Lozano Diz, Z. Pilat, D. Polli, P. Strömert, C. Sturm, R. Vanna, R. Portela. Artificial Intelligence-Powered Raman Spectroscopy through Open Science and FAIR Principles. ACS Nano 2025, 19, 38189–38218.